View unanswered posts    View active topics

All times are UTC - 6 hours





Post new topic Reply to topic  [ 67 posts ] 
Go to page Previous  1, 2, 3, 4, 5  Next

Print view Previous topic   Next topic  
Author Message
Search for:
PostPosted: Sun May 20, 2018 5:57 pm 
Offline
Joined: Wed Jan 07, 2004 12:14 pm
Posts: 392
Location: Charlotte, NC
Am running: VGA compatible controller: NVIDIA Corporation G96 [GeForce 9400 GT] (rev a1)
Driver: x86_64 Kernel Module 340.104 Thu Sep 14 17:13:13 PDT 2017
GCC version: gcc version 5.3.0

Just for grins I back-rev'ed to the Chrome version in the production repo's and was able to run Chrome...

_________________
Backend server - 3.75 TB 3.0ghz dual core 6 gig RAM, nVidia 9400, Gigabyte GA-870A-UD3 MB, 2 HD-5500, 2 HD Homerun dual tuners, 3 frontend machines - LinHES 8.5


Top
 Profile  
 
PostPosted: Sun May 20, 2018 7:47 pm 
Offline
Joined: Thu Sep 27, 2007 5:44 pm
Posts: 459
brfransen, thanks for the rapid response. I apologize, but I decided to abort. I had a clonezilla image backed up this morning and I restored it to get me back to undo the updates. I'm going to look into getting another hard drive so that I can easily switch back and forth between the production repo and testing repo and try again. My notes are as follows:

  • lirc was triple pressing. I tried to update the lircrc file to get this to change in any way. The default button press was set for
    Code:
    repeat = 3
    I tried changing it to
    Code:
    repeat = 1
    I also tried adding
    Code:
    delay = 1
    It did not get any better (or worse). Additionally, it seemed that the first button I pressed on bootup would only double activate, then all subsequent buttons would triple activate. I did check irw and saw the sequence of button press and release (now _EVUP instead of _UP) 3 times (2 times on the first button press). Now that I have restored the previous image, I am checking irw and I see 3 presses followed by one release. I presume the issue lies in the fact that the new lirc adds a release to each press. Thinking about it even further, I think it is a perfect storm that this change happened and I am actually using a Logitech Harmony 700 universal remote which may be the origination of the triple press. I would have to dig out the original streamzap to compare.
  • I prefer to use vim for text editing. It will not load after I update testing. It was due to a shared library (sorry, I did not record the name). Based on your other comments, this probably just needs to be recompiled.
  • In my setup, I've compiled Kodi 18 alpha 1 which is working reasonably well. After the update, there were library issues with this. I wasn't confident that I was going to be able to recompile quickly.

You've obviously put a lot of work into managing all of this and we appreciate what you're doing. Hopefully I can jump back into it next weekend. Leave the lirc issue with me because I have a hunch at this point that it is related to my particular setup.


Top
 Profile  
 
PostPosted: Mon May 21, 2018 9:41 am 
Offline
Joined: Fri Jul 21, 2006 11:12 pm
Posts: 1096
Location: SC
nbdwt73 wrote:
Am running: VGA compatible controller: NVIDIA Corporation G96 [GeForce 9400 GT] (rev a1)
Driver: x86_64 Kernel Module 340.104 Thu Sep 14 17:13:13 PDT 2017
GCC version: gcc version 5.3.0

Just for grins I back-rev'ed to the Chrome version in the production repo's and was able to run Chrome...

If that is from testing repos that should be nvidia 340.106 and gcc 7.3.

I am running the same video drivers. But I don't think that would cause the segfault you are seeing. And since Chrome is a static binary I can't think of any reason why you would have an issue and I wouldn't.


Top
 Profile  
 
PostPosted: Mon May 21, 2018 9:50 am 
Offline
Joined: Fri Jul 21, 2006 11:12 pm
Posts: 1096
Location: SC
knappster:

Try setting repeat and delay to 0 just for testing. This should make it only send 1 per press. I am wondering if maybe the in kernel stuff is not getting disabled for the streamzap. But irw showing 3 seems to indicate that it is lirc doing the repeats.

I rebuilt vim so next update you try it should work.


Top
 Profile  
 
PostPosted: Mon May 21, 2018 11:03 am 
Offline
Joined: Wed Jan 07, 2004 12:14 pm
Posts: 392
Location: Charlotte, NC
I agree - cant think of a reason or find any link. I keep backup harddrive images... I will load the last backup and reload from the test repos again. I did not watch carefully the first time to see if there were any issues with Chrome - I doubt it but worth a try.

_________________
Backend server - 3.75 TB 3.0ghz dual core 6 gig RAM, nVidia 9400, Gigabyte GA-870A-UD3 MB, 2 HD-5500, 2 HD Homerun dual tuners, 3 frontend machines - LinHES 8.5


Top
 Profile  
 
PostPosted: Mon May 21, 2018 11:33 am 
Offline
Joined: Wed Jan 07, 2004 12:14 pm
Posts: 392
Location: Charlotte, NC
Nope - fresh upgrade. First thing after completion was google-chrome... May 21 13:30:10 mythserv kernel: [86542.404797] Chrome_IOThread[3111]: segfault at 20 ip 00007f78daf66060 sp 00007f78c7105690 error 4 in ld-2.26.so[7f78daf5a000+26000]...

_________________
Backend server - 3.75 TB 3.0ghz dual core 6 gig RAM, nVidia 9400, Gigabyte GA-870A-UD3 MB, 2 HD-5500, 2 HD Homerun dual tuners, 3 frontend machines - LinHES 8.5


Top
 Profile  
 
PostPosted: Mon May 21, 2018 11:46 am 
Offline
Joined: Fri Jul 21, 2006 11:12 pm
Posts: 1096
Location: SC
knappster, when you get back to looking at this it might be worth trying this lircd: https://github.com/graysky2/streamzap/b ... emote.conf It is quite a bit different than the LH template for Streamzap.


Top
 Profile  
 
PostPosted: Mon May 21, 2018 2:01 pm 
Offline
Joined: Wed Jan 07, 2004 12:14 pm
Posts: 392
Location: Charlotte, NC
Britney,

I remembered (and just confirmed) that I installed Chromium a long time ago on this machine (still there but unused). Also installed Chrome in later years (default browser). Don't know if it matters (I doubt it) but I uninstalled Chromium and tried again. Still fails but not sure if it is due to left over trash... Thought you might want to know.

knappster, are you having any browser issues?

_________________
Backend server - 3.75 TB 3.0ghz dual core 6 gig RAM, nVidia 9400, Gigabyte GA-870A-UD3 MB, 2 HD-5500, 2 HD Homerun dual tuners, 3 frontend machines - LinHES 8.5


Top
 Profile  
 
PostPosted: Mon May 21, 2018 2:45 pm 
Offline
Joined: Thu Sep 27, 2007 5:44 pm
Posts: 459
nbdwt73 wrote:
Britney,

I remembered (and just confirmed) that I installed Chromium a long time ago on this machine (still there but unused). Also installed Chrome in later years (default browser). Don't know if it matters (I doubt it) but I uninstalled Chromium and tried again. Still fails but not sure if it is due to left over trash... Thought you might want to know.

knappster, are you having any browser issues?


I haven't gotten that far yet. Should have a new hard drive on Wednesday but may not be able to clone it until this weekend. I might perform the updates on my virtual machine in the meantime since it should be easier to backup and restore. With that I can confirm whether chrome works, whether the recompiled vim works, and rebuild kodi 18 alpha 1.


Top
 Profile  
 
PostPosted: Mon May 21, 2018 2:50 pm 
Offline
Joined: Fri Jul 21, 2006 11:12 pm
Posts: 1096
Location: SC
Chromium probably wouldn't even run with other lib issues.

I just tried testing Chrome in a VM and no issues so I think that eliminates it being graphic driver related.

Could you try removing the ~/.config/google-chrome and ~/.cache/google-chrome dirs to see if maybe that fixes it?


Top
 Profile  
 
PostPosted: Tue May 22, 2018 11:26 am 
Offline
Joined: Thu Sep 27, 2007 5:44 pm
Posts: 459
I copied my virtualbox virtual machine to a different computer and ran the testing updates on it and google-chrome launched fine for me. I also tested vim which is now working as well. One thing I will note is that the host computer is running an old nvidia card with the nouveau driver. I don't remember why I'm not using the nvidia driver, but the consequence is that I was unable to enable hardware video acceleration in the virtual machine (likely because it's not enabled on the host machine).

While I never had any google-chrome segfaults when using nvidia-304 drivers with my Geforce 210, I did often have trouble launching it. I see you are on the 340 driver and I don't think it had the same symptoms.

nbdwt73, based on brfransen's comments, is it possible that you only updated extra to testing, but not core? It sounds like your gcc and nvidia driver packages (core) did not update, but google-chrome (extra) did.


Top
 Profile  
 
PostPosted: Thu May 24, 2018 10:52 am 
Offline
Joined: Wed Jan 07, 2004 12:14 pm
Posts: 392
Location: Charlotte, NC
Sorry to have gone silent - Have been dealing with some business issues. So, trying to figure this seg fault issue I decided to try a different approach. My main system has been around since the 7.2 release and simply applied updates since then. Additionally, I run several added things on this box (none of which relate to this problem) so I put a new drive in the box, downloaded 8.5.1 and did a fresh install. Could not get passed the step where the machine reboots after the install completes - in verbose mode it hangs at the time of the initial backend start with a message "unable to start backend" (nothing else) and a restart (not a complete reboot) at the initial LinHes splash screen... over and over.

Tried again with 8.5 - same thing. Then 8.4.3 which went fine... Then simply pacman -Syu from there back to 8.5.1. Now... installed the test repos (knappster, both core and extra are installing correctly) to where I was last weekend. No more seg faults and all seems smooth. Been running for 2 days. Only known issue has to do with Plex but minor and their problem...

_________________
Backend server - 3.75 TB 3.0ghz dual core 6 gig RAM, nVidia 9400, Gigabyte GA-870A-UD3 MB, 2 HD-5500, 2 HD Homerun dual tuners, 3 frontend machines - LinHES 8.5


Top
 Profile  
 
PostPosted: Thu May 24, 2018 1:06 pm 
Offline
Joined: Fri Jul 21, 2006 11:12 pm
Posts: 1096
Location: SC
nbdwt73,

The install failure is concerning. To be clear it sounds like the frontend service which starts Xorg isn't able to start. Can you confirm? If you open another console or ssh in and run xconfig.sh does that help?


Top
 Profile  
 
PostPosted: Thu May 24, 2018 2:06 pm 
Offline
Joined: Wed Jan 07, 2004 12:14 pm
Posts: 392
Location: Charlotte, NC
Tried several things. First, to confirm not a drive problem, installed another with same results. Alt-X at splash screen put me into verbose mode but once there system would no longer respond to keyboard (oddly enough works fine now when I force to verbose at roughly the same place). SSHed in - message below. I am going to try another box (same guts, just different video card) tonight.

May 24 16:37:24 mythserv1 su: (to mythtv) root on console
May 24 16:37:24 mythserv1 acpid: client 3150[0:1000] has disconnected
May 24 16:37:24 mythserv1 acpid: client connected from 3241[0:1000]
May 24 16:37:24 mythserv1 acpid: 1 client rule loaded
May 24 16:37:24 mythserv1 kernel: [ 258.058096] nvidia: disagrees about version of symbol module_layout

_________________
Backend server - 3.75 TB 3.0ghz dual core 6 gig RAM, nVidia 9400, Gigabyte GA-870A-UD3 MB, 2 HD-5500, 2 HD Homerun dual tuners, 3 frontend machines - LinHES 8.5


Top
 Profile  
 
PostPosted: Fri May 25, 2018 9:20 am 
Offline
Joined: Fri Jul 21, 2006 11:12 pm
Posts: 1096
Location: SC
That last line indicates something with the nvidia driver and card. Can you run xconfig.sh to see what driver it chose.


Top
 Profile  
 

Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 67 posts ] 
Go to page Previous  1, 2, 3, 4, 5  Next



All times are UTC - 6 hours




Who is online

Users browsing this forum: No registered users and 1 guest


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Jump to:  
Powered by phpBB® Forum Software © phpBB Group

Theme Created By ceyhansuyu